3DM6

When working in the areas of data and analytics, I have found it consistently helpful over many years to organize activity into one of six components of data-driven decision making; Data, Analysis, Insight, Decision Feedback, Measurement. #3DM6.
Seems too simple and logical to be helpful yet the benefits are significant when ensuring components are present and connected is taken seriously. Combining that framework with the principle of keeping logic visible and working towards a state of continual improvement will help build a data culture that is collaborative, agile, innovative and connected to business value. It will help customers, consumers, and suppliers of data and analytics services to:

• recognize opportunities
• build the business case
• scope the project
• monitor implementation

All that will get you to value from your data sooner with less waste and lower risk.
It could be especially helpful to a C Suite who would like to create that elusive data-driven culture.
Managers who are responsible for delivering value beyond technology, solution architects, project managers, statisticians, data scientists etc. could all benefit.
That was a lot of claims that I would appreciate the chance to discuss and defend. #3DM6.

How to tell Data and Analytics stories to Executives:
At a minimum, ensure you touch on 6 important components of successful data-driven decision making.
#3DM6 #data #analysis #insight #decision #feedback #measurement

A simple example: Associations in transactional data enables a fraud prediction which will improve our decisions to approve credit. We expect a #% reduction in fraud worth $##. We will monitor performance to validate and enable improvement over time.

You could go deeper, but at minimum organizing the story by those components lets execs know why they care, triggers good questions and frames the deeper discussion.

Pet peeve; Being told it is better to dazzle the execs with a fancy visualization and talk in metaphors. It is even more frustrating hearing “told you so” when the dazzle with metaphors strategy works.

Automating decision making is a big deal. When developing algorithms that could become a competitive advantage or major risk, it is especially important the executive or even the board insist on a comprehensive and structured way to understand what is happening at an appropriate level. #3DM6.

A well-defined problem is important for a successful analytics project.

Understanding the decision or action you would like to influence is an important milestone.

Discussions with subject matter experts that focus on decisions and the value of improving them can get you there. A request to improve a prediction that would improve a decision is something a data scientist could really sink their teeth into.

Important inputs to a successful implementation include knowledge of the decision to be influenced, where it fits in the process, downstream impacts, upstream support, education or system modifications needed, and especially who the key stakeholders are.

When using data and analytics to improve decisions and actions, be cautious of adding a sophistication that is not worth the effort or cost.

Complexity makes it more difficult to recognize, diagnose and fix errors. Coordination across disciplines becomes more challenging. Human intuition is less likely to engage when needed and support for the implementation could be lower if the solution is a black box no one understands.

Measured incremental improvement from a base that is understandable and useful for the task at hand is worth considering.
Guard against oversimplifying by ensuring six important components of successful data-driven decision making are present in a state of continual improvement. #3DM6

“Simplicity is the ultimate sophistication”. Leonardo Da Vinci
I think a bit more of Leonardo’s thinking would help. #analytics, #machinelearning, #artificialintelligence, #businessintelligence

Business Subject Matter Experts can be more active in defining opportunities to exploit data and analytics capability, without a data scientist.

Think about the decisions made within your area of expertise or the ones you would like to influence. Determine the value of good choices and which decisions would have the most benefit if improved. Do the calculation, connect to revenue, costs or risks, even if it takes many assumptions.

Most decisions involve predictions. Identify them and identify what data might improve those predictions or measure the outcome. Don’t restrict yourself. There is plenty of time to figure out what is NOT possible, not allowed or not cost effective later, and the restriction is probably temporary.

At this stage, you have identified predictions you would like improved, data you need, and the business case for doing it. Don’t lose that calculation connecting all this activity to business value. It is a key to managing scope and being helpful as the project progresses. Don’t stop referring to it, adjusting it, improving it.

Companies experience challenges getting value from their investments in data and analytics. Large consulting companies recommend that you collaborate and connect to the business. They also recommend being innovative, agile and creative. I’d like to take a shot at providing a level of level of detail on how to achieve that.

It can be quite effective to take a step back from technology and organize the work into logical components. I have always found it very helpful to consider the following components; Data Analysis Insight Decision Feedback and Measurement. Six important components of successful data-driven decision making (3DM6). Data analyzed for insight does not have value until it impacts a decision. Feedback avoid errors and find opportunities to improve. Measuring the overall impact helps justify and manage project scope and objectives.

No rocket science there, but don’t underestimate the value of the approach because it is simple and logical. Being serious about ensuring components are present and connected, combined with keeping logic visible and working towards a state of continual improvement will help build a data culture and directly benefit customers, consumers, and suppliers of data and analytics services. It can help recognize opportunities, build the business case, scope the project and monitor implementation. It can be helpful to a C Suite who would like to create a more data-driven culture, managers responsible for delivering value beyond technology, solution architects, project managers, statisticians, data scientists etc. It can help build a data-oriented culture that is collaborative, creative, agile, and business value oriented.

I just made any claims that probably sound too good to be true? I plan to back up them up in follow up blogs as well as illustrate why this approach, while it does require some determination and belief, is not difficult to implement. Please reach out if you would like to discuss those claims directly.